The host microenvironment has a significant influence on the growth and metastasis of solid tumors. One host cell type, the myoepithelial cell, forms a layer in juxtaposition to normal ductal epithelial cells in health and in pre/early-cancerous disease states such as ductal carcinoma in situ (DCIS). We have seen that the myoepithelial cells inhibit angiogenesis, the formation of new capillaries. This angiogenesis-suppressive phenotype is reflected in the preponderance of expressed angiogenic inhibitors to angiogenic factors, and the fact that the myoepithelial cell lines inhibit angiogenesis in a paracrine fashion in in vitro and in vivo experiments. Furthermore, we have observed that the myoepithelial xenografts exhibit a growth pattern that is a angiogenesis-independent, since they grow in nude mice but contain little evidence of murine angiogenesis within their matrix. We therefore propose to identify angiogenesis-independent genes expressed by the myoepithelial xenografts by using an animal selection model. We will also aim to characterize these genes and their protein products with angiogenesis assays and animal tumor and transgenic models. The results of this proposal may contribute significantly to the understanding of the regulatory mechanisms of tumor angiogenesis.